THz spectrum of monodeuterated methane

ARTICLE IN PRESS

Journal of Quantitative Spectroscopy & Radiative Transfer 109 (2008) 580–586 www.elsevier.com/locate/jqsrt

THz spectrum of monodeuterated methane Valerio Lattanzia,b, Adam Waltersb, John C. Pearsonc, Brian J. Drouinc, a

Department of Physics, University ‘‘La Sapienza’’, P.le Aldo Moro 2, Rome, Italy Centre d’Etude Spatiale des Rayonnements, 9, avenue du Colonel Roche—Boite postale 4346 31028 Toulouse Cedex 4, France c Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109-8099, USA

b

Received 5 July 2007; received in revised form 6 September 2007; accepted 9 September 2007

Abstract We report new measurements of the rotational spectrum of monodeuterated methane ðCH3 DÞ in the range of 690–1200 GHz which allow for an accurate prediction of all lines in the range of the high-resolution spectrometer of the Herschel Space Observatory. Comparison is also made with the previous analysis based on infrared combination differences. Three lines of 13 CH3 D were measured in natural abundance. r 2007 Elsevier Ltd. All rights reserved. PACS: 98.35.Bd; 33.20.-t; 33.20.Bx; 33.20.Sn Keywords: Methane; Line positions; Deuterated methane; Interstellar spectra

1. Introduction We report extended measurements of the rotational spectrum of monodeuterated methane (CH3 D). This species is important for the study of planetary atmospheres and of the interstellar medium (ISM). Since the first detection of CH3 D in the atmosphere of Jupiter [1], the deuterium-to-hydrogen (D/H) ratio has played a major role in the study of the origin and evolution of planetary atmospheres (e.g. [2]). The large mass difference between hydrogen and its isotope deuterium induces significant differences in both the thermodynamics and the kinetics of various processes, such as thermal escape, chemical reactivity and condensation, resulting in an isotopic fractionation. Therefore observed D/H ratios act as tracers of the physical and chemical history of planetary atmospheres as they evolve from the initial interstellar ices and gases. CH3 D has been observed on Jupiter [3,4], Saturn [5,6], Titan [7–10], Uranus [11,12] and Neptune [13] by, for example, ISO and the IRIS instrument on Voyager. The weaker rotational spectrum of CH4 has been observed with the CIRS instrument on Cassini and has been used to determine the methane abundance in Saturn and Titan: see Ref. [14] (Saturn) and Ref. [15] (Titan). Recently the Cassini-CIRS instrument has also reported measurements of the isotopic methane species CH3 D [16] and 13 CH3 D [17] on Titan.

Corresponding author. Tel.: +1 818 393 6259; fax: +1 818 354 5148.

E-mail address: [email protected] (B.J. Drouin). 0022-4073/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jqsrt.2007.09.002

ARTICLE IN PRESS V. Lattanzi et al. / Journal of Quantitative Spectroscopy & Radiative Transfer 109 (2008) 580–586

581

The deuterium present in the atmospheres of the giant planets is thought to be representative of the protosolar composition, a quantity of high interest in astrophysics [18]. The D/H ratio of methane has also been used to estimate the formation temperature of cometary molecules [19]. The authors used this determination as an evidence that our Sun was born in a warm cloud of 30 K rather than a cold one at 10 K. It should be noted that this conclusion was based on the non-detection of CH3 D in the near-infrared. The launch of the Herschel Space Observatory in 2008 will enable measurements of astrophysical spectra in the far-infrared that are not possible from earth-based observatories because of the atmospheric absorption. In particular the HIFI instrument will allow for very high-resolution spectral studies to be taken over a wide spectral range from around 500–1900 GHz, increasing the region in which new spectral lines can be searched for and allowing wide and long scans unaffected by local weather conditions. Laboratory spectral data for CH3 D in this region are lacking and only three precisely measured rotational transitions, all below 500 GHz, have been reported. Chemical models predict methane to be one of the most abundant polyatomic species in dense interstellar clouds and it is thought to have enhanced abundance in hot, dense gas such as the Orion hot core region because of evaporation from grain mantles. Although its rovibrational spectrum has been detected toward IRC +10216 [20], vibrational excitation is limited in interstellar molecules, especially in regions of extended gas that dominate molecular clouds, since even toward star-forming objects temperatures are typically only around 10–100 K. Hence, infrared spectra must be measured in absorption. Furthermore, vibrational spectra cannot be used to study far inside clouds containing dust since the latter absorbs and scatters infrared radiation. However, far-infrared radiation can penetrate throughout the cloud. Since it is completely symmetric CH4 has no permanent dipole moment and is not easily identified in the ISM from its rotational spectrum. The dipole moment of CH3 D that arises from the isotopic substitution is small but non-zero (6  103 D) [21,22]. This being measured via direct absorption methods in the THz spectrum at low resolution and using the electric resonance spectrum in the first two rotational states. The dipole was later shown to have a significant rotational dependence [23]. CH4 also has a centrifugally induced dipole moment that is around a 1000 times weaker than the isotopically induced electric dipole of CH3 D [24]. The cosmic elemental D/H ratio is expected to be around 1:522:3  105 (see, for example, Ref. [25]). However, for molecular species the fractionation ratio, defined as the ratio of the column density of a deuterated molecule to its hydrogen counterpart, is found to be up to five orders of magnitude higher than this elemental abundance ratio [26]. The deuterium enhancement results from chemical processes, which can involve both gas-phase and surface reactions. Hence it is not clear whether CH3 D or CH4 is most likely to be first detected. An attempt to identify CH3 D in Orion proved inconclusive [27], giving an upper limit to the column density of the order of 1018 cm1 . The ALMA interferometer which will come progressively into service at the end of this decade will give not only high spatial resolution allowing searches to concentrate on specific areas of supposed high concentration but also highly increased sensitivity due to the cumulative surface area of the array of telescopes used (up to 64). Previous laboratory measurements of the rotational spectrum of deuterated methane have been limited by available technology. Until recently only three lines had been measured, the Jð0 ! 1Þ, by Pickett et al. [28], and the two K-components of Jð1 ! 2Þ by Womack et al. [27], who also measured the first transition as well. The predictive analysis available from ground-state combination differences (GSCDs) [29] and the available pure rotational data have been the basis of the JPL catalog compilation. The present work extends the basis of high precision pure rotational data. We now report 12 new measured rotational frequencies in the range of 697–1162 GHz, completing measurements of the rotational spectrum up to J 0K 0 ¼ 54 . 2. Experimental The measurements were carried out at JPL. The spectrometer system has been described earlier [30,31]. Briefly, a millimeter-wave module, with 10–100 mW of output power, and a series of commercial (Virginia Diodes) and JPL built multiplier chains were used to produce THz radiation. The radiation source was a sweep synthesizer phase-locked to a frequency standard with a precision of one part in 1012 ; so the frequency

ARTICLE IN PRESS 582

V. Lattanzi et al. / Journal of Quantitative Spectroscopy & Radiative Transfer 109 (2008) 580–586

21 -> 31

20 -> 30

22 -> 32 Prediction

697650

697684

697718

697752

697786

697820

Frequency (MHz) 33 -> 43

32 -> 42

31 -> 41

30 -> 40

Prediction

929902

929972

930041 930111 Frequency (MHz)

930180

930250

Fig. 1. Extract of two scans with (top) and without (bottom) numerical derivative filter for removing the baseline. Predictions of the transitions from the microwave-fit are also shown. The two additional lines probably result from trace contamination; ethanol was used to clean the cell but there is no obvious assignment.

error depends entirely on determining the line center. Tone burst modulation was employed and the cell was used in a double-pass arrangement, giving 2 m total length. The detector was a Si bolometer cooled to 2.1 K with pumped 4 He liquid. In order to search for lines we first made predictions using the constants obtained by Ulenikov et al. [29]. A scan was made for each successive J value, including all K-components. Conditions were optimized in the 929 GHz range, with a good compromise between linewidth and signal-to-noise (S/N) obtained at 280 m Torr. The cell was at room temperature. In the 929 GHz range, where the strongest source was available, four identical scans were co-added and gave sufficient S/N for precise frequency measurements. In the other frequency regions, longer (overnight) observation times were used (697 GHz—50 scans and 1162 GHz—97 scans co-added). Two spectral scans are shown as examples in Fig. 1. 3. Analysis In the attempt to identify monodeuterated methane in Orion, Womack et al. [27] also provided new measurements for the Jð1 ! 2Þ and Jð0 ! 1Þ transitions which allowed three molecular parameters, B0 , DJ and DJK to be determined. Starting from these data a global microwave analysis was performed, including our measurements, from Jð2 ! 3Þ to Jð4 ! 5Þ, using Pickett’s [32] SPFIT and SPCAT programs. All the

ARTICLE IN PRESS V. Lattanzi et al. / Journal of Quantitative Spectroscopy & Radiative Transfer 109 (2008) 580–586

583

transitions included in the microwave fit and the corresponding residuals (observed  calculated) are listed in Table 1. A 50 kHz uncertainty was assumed for all our measurements, except the 40 ! 50 . In this case the line is too close to a stronger transition of another unidentified species. CH3 D is a C 3v prolate symmetric top. The pure rotational Hamiltonian leads to the following energy levels [33]: EðJ; KÞ=h ¼ B0 ½JðJ þ 1Þ  K 2  þ A0 K 2  D0J J 2 ðJ þ 1Þ2  D0JK JðJ þ 1ÞK 2  D0K K 4 þ H 0J J 3 ðJ þ 1Þ3 þ H 0JK J 2 ðJ þ 1Þ2 K 2 þ H 0KJ JðJ þ 1ÞK 4 þ H 0K K 6 .

ð1Þ

Due to the selection rule for the allowed rotational transitions DJ ¼ 1, DK ¼ 0, the pure-K parameters such as A0 , D0K and H 0K cannot be obtained by the analysis of a pure rotational spectrum. They can, however, be derived by combination differences from rovibrational spectra and ‘‘perturbation allowed’’ transitions, with DJ ¼ 0 and DK ¼ 1; 3. The K ¼ 3 energy splitting of a generic rotational level J can be derived from the off-diagonal elements of the Hamiltonian, and is DJ3 ¼ 2JðJ þ 1Þ½h03 þ hJ3 JðJ þ 1Þ þ   F ðJÞ,

(2)

where F ðJÞ ¼ ½JðJ þ 1Þ  2½JðJ þ 1Þ  6.

(3)

Since it depends on the difference between the energy of the J and J þ 1 levels, this splitting is only observed at higher values of J. Using the h3 constants reported by [29], the separation between the two K ¼ 3 states is smaller than 100 kHz for J 00 p5. For the K ¼ 3 transitions only a single line was observed with a width of around 2.5 MHz comparable to the other lines. Hence the splittings appear completely unresolved and no improvement of the h03 constant and its centrifugal distortional corrections have been provided by our work. In Table 2 we report the molecular parameters determined by our analysis, compared with the analysis of Womack et al. [27]. We have improved the three parameters already determined; B0 is slightly better determined, the uncertainty on DJ is reduced by a factor of three and that of DJK by 25. We also constrained for the first time with microwave data three sextic constants. Ulenikov et al. [29] present two Hamiltonians for the ground state of CH3 D that include rotational operators up to eighth order. The models were adequate to reproduce the GSCDs to within 5% of the Table 1 Measured transition used for the microwave fit J 00

K 00

J0

K0

Frequency

Residual

Uncertainties

1 2 2 3 3 3 4 4 4 4 4 5 5 5 5 5 5

0 1 0 2 1 0 3 3 2 1 0 4 3 3 2 1 0

0 1 1 2 2 2 3 3 3 3 3 4 4 4 4 4 4

0 1 0 2 1 0 3 3 2 1 0 4 3 3 2 1 0

232 644:301a 465 235:540a 465 250:691a 697690.6269 697758.7502 697781.4559 929926.5710 929926.5710 930077.7603 930168.4642 930198.6776 1161 860.7890 1162 125.1934 1162 125.1934 1162 313.6641 1162 426.6568 1162 464.4099

37 3 9 4 9 4 16 16 40 11 22 12 36 36 25 33 72

75 75 75 50 50 50 50 50 50 50 50 50 50 50 50 50 75

Frequencies are in MHz; residuals and uncertainties in kHz. For assigned uncertainties see text. a From Womack et al. [27].

ARTICLE IN PRESS 584

V. Lattanzi et al. / Journal of Quantitative Spectroscopy & Radiative Transfer 109 (2008) 580–586

Table 2 Molecular constants of CH3 D Constant

Microwave (MHz)

Ref. [22] (MHz)

Ref. [24] (MHz)

A B DJ DJK DK H J  103 H JK  103 H KJ  103 H K  103 LJJK  106 LJKK  106 h~3  106 ~  101

– 116 325.2840(145) 1.57594(96) 3.79343(226) – 0.0280(191) 0.583(60) 0.490(90) – – – – –

– 116 325.309(12) 1.5796(21) 3.79(3) – – – – – – – – –

157 415.656(53) 116 325.271(27) 1.57781(12) 3.79039(46) 2.367339(47) 0.04301(16) 0.36410(19) 0.2015(17) 0.0493(15) 0.0355(22) 0.0447(51) 0:9744ð65Þa 0:96a

MWRMS sRMS b

0.029 0.519

– –

0.194 3.77

Standard errors (1s) in parentheses refer to the least significant digits. MWRMS are the averages of the least-squares residuals of the fits. a~ h3 and ~ are defined in Ref. [29], ~ is a fixed parameter. b Dimensionless, weighted standard deviation of each fit.

Table 3 Measured transitions of

13

CH3 D

J 00

K 00

J0

K0

Frequency

Residual

Predicted [33]

4 4 4

3 1 0

3 3 3

3 1 0

929132.1265 929376.0805 929406.3318

13 120 106

929130.7 929375.6 929406.2

Frequencies are in MHz; residuals in kHz. The assigned uncertainties is 200 KHz for all the lines.

experimental uncertainty. Without published values for the GSCDs we were unable to make a direct comparison of our measurements to this data set. Instead, the calculated energy levels listed in Ulenikov et al. [29] were recalculated with SPFIT/SPCAT to ensure an identical model could be created; this procedure was done for both Model 1 and Model 2 (see Table 2) and reproduced the energy values to 4  106 and 7  106 cm1 , respectively. This level of difference is expected for reproduction of a calculated data set that is truncated in the 106 cm1 digit. Using these exact models the new microwave data set was evaluated to have the rms deviations of 189 or 193 kHz, respectively. The predicted uncertainties using Model 2 are 150–212 kHz. This is approximately 3.7 times the estimated uncertainty in the new microwave data and is a direct measurement of the relative qualities of the data sets. Three transitions of 13 CH3 D were also detected in natural abundance and are reported in Table 3. Our measurements are compared with the predictions based on the A0 , B0 , D0K and D0JK constants reported by Ulenikov et al. [38] from infrared combination differences. 4. Conclusion We have performed an extended analysis up to the THz region of deuterated methane. Methane is a key molecule for astrophysics and for understanding chemical evolution of interstellar medium. Theoretical works [34–36] and laboratory measurements [37] have predicted and observed pure rotational spectra of T d symmetry spherical tops such as methane; however, the rotational transitions that occur are weak and have not yet been detected in the ISM. Since deuterium enrichment may make CH3 D more easily detectable than

ARTICLE IN PRESS V. Lattanzi et al. / Journal of Quantitative Spectroscopy & Radiative Transfer 109 (2008) 580–586

585

CH4 , it is important to have a precisely determined microwave spectrum for both. The Atacama Large Millimeter Array (ALMA) with unprecedented spatial resolution and sensitivity could allow spatially resolved determination of the D=H ratio. The aim of this work is principally to provide accurate line frequencies for astrophysics and planetary science in the range to be opened by the next generation of radiotelescopes and far-infrared instruments. The A0 , D0K and H 0K constants could not be determined from the new microwave data alone. For completeness we hence also compared to previous data from infrared combination differences [29]. However, these constants should not be necessary for accurate predictions for radioastronomy. Rotational lines of 13 CH3 D have been reported for the first time. Acknowledgments V.L. is a student financed by the Universite´ Franco-Italienne. His visit to JPL for the measurements was financed by the Observatoire Midi-Pyre´ne´es, the PCMI and by an ATUPS grant from Paul Sabatier University. A portion of this research was performed at the Jet Propulsion Laboratory, California Institute of Technology, under contract with National Aeronautics and Space Administration. Financing by the EU Molecular Universe Training Network also allowed travel for discussion on the analysis with B.J.D. during a visit to Europe. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NASA. References [1] Beer R, Taylor FW. Abundance of CH3 D and D/H ratio in Jupiter. Astrophys J 1973;179:309–28. [2] Fouchet T, Lellouch E. Vapor pressure isotope fractionation effects in planetary atmospheres: application to deuterium. Icarus 2000;144:114–23. [3] Knacke RF, Kim SJ, Ridgway ST, Tokunaga AT. The abundances of CH4 , CH3 D, NH3 , and PH3 in the troposphere of Jupiter derived from high-resolution 110021200 cm1 spectra. Astrophys J 1982;262:388–95. [4] Kunde V, Hanel R, Maguire W, Gautier D, Baluteau JP, Marten A, et al. The tropospheric gas-composition of Jupiters north equatorial belt (NH3 , PH3 , CH3 D, GeH4 , H2 O) and the Jovian D/H isotopic ratio. Astrophys J 1982;263:443–67. [5] Noll KS, Larson HP. The spectrum of Saturn from 1990 to 2230 cm1 —abundances of AsH3 , CH3 D, CO, GeH4 , NH3 , and PH3 . Icarus 1991;89:168–89. [6] Courtin R, Gautier D, Marten A, Bezard B, Hanel R. The composition of Saturn’s atmosphere at northern temperate latitudes from Voyager IRIS spectra—NH3 , PH3 , C2 H2 , C2 H6 , CH3 D, CH4 , and the Saturnian D/H isotopic ratio. Astrophys J 1984;287:899–916. [7] Penteado PF, Griffith CA, Greathouse TK, de Bergh C. Measurements of CH3 D and CH4 in Titan from infrared spectroscopy. Astrophys J 2005;629:L53–6. [8] Coustenis A, Salama A, Schulz B, Ott S, Lellouch E, Encrenaz Th, et al. Titan’s atmosphere from ISO mid-infrared spectroscopy. Icarus 2003;161:383–403. [9] Mousis O, Gautier D, Coustenis A. The D/H ratio in methane in Titan: origin and history. Icarus 2002;159:156–65. [10] Coustenis A, Bezard B, Gautier D. Titan atmosphere from Voyager infrared observations II. The CH3 D abundance and D/H ratio from the 90021200 cm1 spectral region. Icarus 1989;82:67–80. [11] Lutz BL, de Bergh C, Maillard JP, Owen T, Brault J. On the possible detection of CH3 D on Titan and Uranus. Astrophys J 1981;248:L141–5. [12] de Bergh C, Chauville J, Lutz BL, Owen T, Brault J. Monodeuterated methane in the outer solar-system. II Its detection on Uranus at 1.6 microns. Astrophys J 1986;311:501–10. [13] de Bergh C, Lutz BL, Owen T, Maillard JP. Monodeuterated methane in the outer solar-system. IV Its detection and abundance on Neptune. Astrophys J 1990;355:661–6. [14] Flasar FM, Achterberg RK, Conrath BJ, et al. Temperatures, winds and composition in the Saturnian system. Science 2005;307:1247–51. [15] Flasar FM, Achterberg RK, Conrath BJ, et al. Titan’s atmospheric temperatures, winds, and composition. Science 2005;308:975–8. [16] Coustenis A, Achterberg RK, Conrath BJ, et al. The composition of Titan’s stratosphere from Cassini/CIRS mid-infrared spectra. Icarus 2007;189(1):35–62. [17] Be´zard B, Nixon CA, Kleiner I, Jennings DE. Detection of 13 CH3 D on Titan. Icarus, 2007, doi:10.1016/j.icarus.2007.06.004. [18] Lutz BL. Current studies of CH3 D: Ye olde line drive. NASA Goddard space flight center vibrational-rotational spectry. Planet Atmos 1982;2:599–610. [19] Kawakita H, Watanabe JI, Furusho R, Fuse T, Boice DC. Nuclear spin temperature and deuterium-to-hydrogen ratio of methane in comet C/2001 Q4 (NEAT). Astrophys J 2005;623:L49–52. [20] Hall DNB, Ridgway ST. Circumstellar methane in the infrared-spectrum of IRC þ10 216. Nature 1978;273:281–2.

ARTICLE IN PRESS 586 [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38]

V. Lattanzi et al. / Journal of Quantitative Spectroscopy & Radiative Transfer 109 (2008) 580–586 Ozier I, Ho W, Birnbaum G. Pure rotational spectrum and electric dipole moment of CH3 D. J Chem Phys 1969;51:4873–80. Wofsy SC, Muenter JS, Klemperer W. Hyperfine structure and dipole moment of CH3 D. J Chem Phys 1970;53:4005–14. Watson JKG, Takami M, Oka T. Rotational dependence of the dipole-moment of CH3 D. J Chem Phys 1979;70:5376–80. Ozier I. Ground-state electric dipole moment of methane. Phys Rev Lett 1971;27:1329–32. Roueff E, Herbst E, Lis DC, Philips TG. The effects of an increased elemental D/H ratio on deuterium fractionation in the cold interstellar medium. Astrophys J 2007;661:L159–62. Vastel C, Phillips TG, Caselli P, Ceccarelli C, Pagani L. Deuterium enhancement in Hþ 3 in pre-stellar cores. Philos Trans R Soc London A 2006;364:3081–90. Womack M, Apponi AJ, Ziurys LM. Search for interstellar CH3 D: limits to the methane abundance in Orion-KL. Astrophys J 1996;461:897–901. Pickett HM, Cohen EA, Phillips TG. Deuterated methane—laboratory rotational spectrum and upper limits for its abundance in OMC-1. Astrophys J Lett 1980;236:L43–4. Ulenikov ON, Onopenko GA, Tyabaeva NE, Schroderus J, Alanko S. On the rotational analysis of the ground vibrational state of CH3 D molecule. J Mol Spectrosc 1999;193:249–59. Drouin BJ, Maiwald FW, Pearson JC. Application of cascaded frequency multiplication to molecular spectroscopy. Rev Sci Instrum 2005;76:Art. #093113. Maiwald F, Drouin BJ, Pearson JC, Mehdi I, Lewen F, Endres C, et al. Planar diode multiplier chains for THz spectroscopy. In Proceedings of 16th international symposium space THz technology. 2005. Pickett HM. The fitting and prediction of vibration–rotation spectra with spin interactions. J Mol Spectrosc 1991;148(2):371–7, see also hhttp://spec.jpl.nasa.govi. Gordy W, Cook RL. Microwave molecular spectroscopy. 2nd ed. New York: Wiley; 1984. Fox K. Theory of pure rotational transitions in vibronic ground state of methane. Phys Rev Lett 1971;27:233–6. Watson JKG. Forbidden rotational spectra of polyatomic molecules. J Mol Spectrosc 1971;40:536–44. Oldani M, Andrist M, Bauder A, Robiette AG. Pure rotational spectra of methane and methane-D4 in the vibrational ground-state observed by microwave Fourier-transform spectroscopy. J Mol Spectrosc 1985;110:93–105. Holt CW, Gerry MCL, Ozier I. Microwave-spectrum of 12 CH4 in ground vibronic state. Phys Rev Lett 1973;31:1033–6. Ulenikov ON, Onopenko GA, Tyabaeva NE, Anttila R, Alanko S, Schroderus J. Rotational analysis of the ground state and the lowest fundamentals n3 , n5 , and n6 of 13 CH3 D. J Mol Spectrosc 2000;201:9–17.